Turkish Journal of Zoology Turk J Zool (2016) 40: 749-757 http://journals.tubitak.gov.tr/zoology/ © TÜBİTAK Short Communication doi:10.3906/zoo-1510-22

CHD : a reliable marker for populations and phylogenetic analysis? Case study of the superfamily Sylvioidea (Aves: Passeriformes)

1,2 1 1 1 1, Mitică CIORPAC , Radu Constantin DRUICĂ , Gogu GHIORGHIȚĂ , Dumitru COJOCARU , Dragoș Lucian GORGAN * 1 Department of Biology, Faculty of Biology, “Alexandru Ioan Cuza” University of Iasi, Iasi, Romania 2 Interdisciplinary Research Department - Field Science, “Alexandru Ioan Cuza” University of Iasi, Iasi, Romania

Received: 07.10.2015 Accepted/Published Online: 09.04.2016 Final Version: 24.10.2016

Abstract: The chromo--DNA binding (CHD) genes are widely used markers for sex determination in and provide a rapid and low-cost method with applicability to a large number of taxa. A good phylogenetic marker displays highly conserved domains and a slow evolution rate, proprieties that CHD genes seem to have. This gives rise to the following question: is the CHD a reliable marker for phylogenetic and bird population analysis? The aim of this study is to investigate whether the CHD gene is a reliable tool for molecular phylogeny and for population analysis within the superfamily Sylvioidea. Our results reveal that CHD genes are good markers for these two analyses, even better than myoglobin.

Key words: CHD genes, cytochrome b, myoglobin, Sylvioidea

Sex identification in birds based on their external the design of universal primers for birds’ sex determina- morphology is difficult (Griffiths et al., 1998), as they are tion (Griffiths et al., 1998; Kahn et al., 1998; Fridolfsson mostly sexually monomorphic (Ong and Vellayan, 2008). and Ellegren, 1999). With the discovery of the chromo-helicase-DNA binding Genetic distances are a frequently used tool to identify protein (CHD) gene (Griffiths and Tlwarl, 1995) in the and assess the species status of closely related taxa (Wesson avian sex (Ellegren, 1996), molecular DNA et al., 1993; Hung et al., 1999; Burbrink et al., 2000; Bradley noninvasive sexing methods, such as the analysis of feath- and Baker, 2001; Cagnon et al., 2004; Parkin et al., 2004; ers, became possible and generated applicability to wildlife Olsson et al., 2005; Newman et al., 2012). This way, genetic DNA forensics (An et al., 2007). The avian CHD1 genes distances are frequently compared between different stud- belong to a family composed of a chromatin organization ies, even if different genetic markers are involved (Helbig modifier (An et al., 2007) domain, an SNF2-related heli- et al., 1995; Baker et al., 2003; Johnson and Cicero, 2004; case/ATPase domain, and a DNA binding domain; thus, Zhang et al., 2007). Fast-evolving parts, like mitochondrial the acronym CHD stands for these (Fridolfsson and El- sequences (Heidrich et al., 1998) or nuclear introns, are legren, 1999). CHD genes have an important role in the suitable for resolving young evolutionary relationships, for avian , as in other organisms, due to their involve- example those between species, whereas older relation- ment in in the control of transcrip- ships are better analyzed with more conservative genes like tion elongation (Simic et al., 2003). This gene has two in- nuclear exons (Lin and Danforth, 2004). Mostly, a com- trons with different lengths for the Z and W chromosome, bination of the two is used to target different parts of a allowing the discrimination between the amplicons of the . Z and W by gel electrophoresis (Dubiec For the superfamily Sylvioidea, in addition to mito- and Zagalska-Neubauer, 2006). Being a functional part of chondrial genes such as cytochrome b, control region, the DNA and with a very slow evolution rate, the CHD or ND2 (NADH-ubiquinone oxidoreductase chain 2), gene is highly conserved, even among distant species. The a number of nuclear genes are used for molecular phy- alignment between the CHD-W sequence in birds and the logeny, genes like fibrinogen beta chain intron 5 (FGB), CHD sequence in mice does not include any gaps, except glyceraldehyde-3-phosphate dehydrogenase (GAPDH), for two 130- and 175-bp deletions (Vucicevic et al., 2013). lactate-dehydrogenase B intron 3 (LDHB), ornithine-de- Its high degree of conservation across species has led to carboxylase exon 6–8, intron 7 (ODC1), recombination

* Correspondence: [email protected] 749 CIORPAC et al. / Turk J Zool activation gene 1 (RAG1), and the widely used myoglobin A total of 21 sequences belonging to 7 taxa, with 6 from intron 2 (myo) (Fregin et al., 2009). Considering this, the the superfamily Sylvioidea (Table 1), were used to assess aim of this study is to investigate whether the CHD gene the reliability of the CHD-Z gene in phylogenetic analy- is a reliable tool for molecular phylogeny and population sis. The cytochrome b (cytb [1012 bp; 337 variable sites, of analysis of the superfamily Sylvioidea. which 176 were informative]), myoglobin intron 2 (myo Blood samples from three species of the genus Ac- [656 bp; 207 variable sites without gaps, of which 176 were rocephalus, A. schoenobaenus, A. arundinaceus, and A. informative]), and chromo-helicase-DNA-binding protein scirpaceus, were collected from five different bird ringing gene (CHD-Z exon and intron [326 bp; 62 variable sites camps in Larga Jijia (Iasi County, Romania) during 2010 without gaps, of which just 13 were informative]) sequenc- and stored in Queen’s Lysis Buffer (Seutin et al., 1991). The es were aligned and concatenated using MEGA 6 (Tamura genomic DNA was isolated and purified using the DNA et al., 2013), defining 3 data sets for the same taxa: 1) cytb IQ System (Promega, Madison, WI, USA), and quanti- sequences; 2) cytb and CHD-Z sequences concatenated; 3) fied through spectrometry and electrophoresis. The CHD cytb and myo sequences concatenated. genes were amplified using PCR methods with one pair The corrected HKY (Hasegawa et al., 1985) distance of specific primers: P2 5’-TCT GCA TCG CTA AAT CCT between taxa for each gene was obtained using PAUP TT-3’ (Griffiths and Tlwarl, 1995) and P8 5’-CTC CCA v4.0b10 (http://paup.sc.fsu.edu/). The optimal substitu- AGG ATG AGR AAY TG-3’ (Griffiths et al., 1998). The tion model was selected according to the Akaike infor- PCR reaction was carried out in 25 µL of total volume mation criterion (Akaike, 1974) using three substitution containing GoTaq Green Master Mix (Promega), prim- schemes (+F; +I; +G, 4 nCat) and 8 different topologies ers (0.2 µM final concentration), DNA template (~12.5 in jModelTest v2.1 (Guindon and Gascuel, 2003; Darriba µg), and nuclease-free water up to the final volume. The et al., 2012). The base composition heterogeneity and chi- amplification was conducted in a SensoQuest Labcycler square (χ2) test of deviation from base composition homo- (SensoQuest GmbH, Göttingen, Germany) under the fol- geneity of variable sites were calculated in PAUP v4.0b10 lowing cycling conditions: 1.5 min of initial denaturation (http://paup.sc.fsu.edu/). Furthermore, the relative com- at 95 °C, followed by 30 cycles of denaturation at 95 °C for position variability (RCV) was computed as a summary 30 s, primer alignment at 48 °C for 45 s, and elongation at statistic for each marker, following the method of Phillips 72 °C for 45 s, followed by a final elongation step at 72 °C and Penny (2003). for 5 min, in accordance with Griffiths et al. (1998). A 3% The maximum likelihood (Baker et al., 2003) tree was agarose gel was run for the PCR products and visualized designed in SeaView (Gouy et al., 2010) using PhyML in UV light. The distinction between males and females (Guindon et al., 2010) and was visualized in DensiTree was made by the presence of one band for males and two v2.1.11 (Bouckaert, 2010). Bayesian inference (BI) analy- bands for females. sis was conducted in BEAST v1.8.0 [Bayesian Evolutionary

Table 1. Phylogenetic analysis data set.

GenBank Acc. No. Taxon Family cytb CHD-Z myo Acrocephalus_sechellensis Acrocephalidae AJ004284 1 AM180353 2 FJ883122 3 Phylloscopus_collybita Phylloscopidae HQ608821 4 AF355151 5 DQ125966 6 Phylloscopus_trochilus Phylloscopidae Z73492 7 AF355150 5 AY887719 16 Pycnonotus_sinensis Pycnonotidae FJ487714 17 EF582413 18 GQ242100 8 Pycnonotus_taivanus Pycnonotidae NC013483 9 EU204971 9 GQ242099 8 Parus_atricapillus Paridae AF347937 10 DQ068391 11 KF183723 12 Gallus_gallus out-group AP003319 13 AF006659 14 NM001167752 15

1Heidrich et al. (1998); 2Dawson et al. (2005); 3Fregin et al. (2009); 4Lei et al. (2010); 5Bensch et al. (2006); 6Fuchs et al. (2006); 7Helbig et al. (1995); 8Zuccon and Ericson (2010); 9Chang et al. (2010); 10Gill et al. (2005); 11Harvey et al. (2006); 12Johansson et al. (2013); 13Nishibori et al. (2005); 14Griffiths et al. (1998);15 Boardman et al. (2002); 16Alstrom et al. (2006); 17Lohman et al. (2010); 18Chang et al. (2008).

750 CIORPAC et al. / Turk J Zool

Analysis Sampling Trees (Drummond et al., 2012)] using A. schoenobaenus, a 166.66% sex ratio from a sample size the substitution model selected by the AIC for each gene of 32 was identified, which suggests sex disequilibrium and a Relaxed Clock: Uncorrelated Log-normal (Drum- within the population, as the male percentage is 62.50%. mond et al., 2006) model with a tree prior This sex ratio trend was constant during the sampling setup to the : Yule Process (Yule, 1925; Gern- dates (Figure 1). A 42.85% sex ratio was observed for A. hard, 2008) model, applied for a UPGMA starting tree. arundinaceus in a sample size of 20 individuals, with 30% One independent Markov chain Monte Carlo simulation males and a constant trend during the sampling campaign. of 10 million iterations for each data set was run through A similar situation was observed for A. scirpaceus, with a CIPRES Science Gateway v.3.3 (Miller et al., 2010). Con- higher sex disequilibrium and a sex ratio of 311.11% in a vergence for the posterior distributions of the parameter sample size of 37 individuals with 75.68% males. estimates was checked using Tracer v1.5 (http://tree.bio. Taking into account the possibility of hybridization be- ed.ac.uk/software/tracer/), and then the summary maxi- tween A. arundinaceus and A. scirpaceus, identified across mum clade credibility (MCC) tree was computed us- Europe, the correlation degree between these two higher ing TreeAnnotator. The MCC trees were visualized and sex disequilibria was tested based on a correlation test that graphically edited in FigTree v1.4.0 (http://tree.bio.ed.ac. showed a strong positive level of correlation (0.81) for the uk/software/figtree/). Finally, the posterior marginal like- males and females of both species and a negative correla- lihoods, treeness index (Phillips and Penny, 2003), Colless tion between the males and females of different species. tree imbalance (Colless, 1982), and mean posterior prob- In the world of birds, quite often both sexes have simi- ability for each tree were used for model comparisons and lar phenotypic traits, and even an experienced ornitholo- identification of the best tree model fit. gist may have difficulty in identifying an ambiguous in- The CHD gene is an easy and reliable tool in bird sex dividual (Dubiec and Zagalska-Neubauer, 2006). Other determination, defining the differences even from the size characteristics, such as sexually monomorphic colors for of the PCR products. This gene presents two alleles of dif- around 60% of all songbird species (Price and Birch, 1996) ferent sizes; the allele from the W chromosome is bigger and the morphological uniformity of nestlings, increase than that of the Z chromosome, allowing for discrimina- the difficulty of taxonomical classification. Therefore, the tion between genera. Sex determination using the CHD utility of the CHD-based technique in molecular sexing gene was conducted for three species of Acrocephalus: A. of bird species is unquestionable, having been proven by schoenobaenus, A. arundinaceus, and A. scirpaceus. For numerous studies. Furthermore, it is a fast and accurate

Figure 1. Sex distribution for A. arundinaceus, A. scirpaceus, and A. schoenobaenus, shown by sampling date.

751 CIORPAC et al. / Turk J Zool method, but at the same time easy to perform and rela- ily distances showed a minimum value for Acrocephali- tively cheap. Furthermore, numerous bird species or pop- dae/Phylloscopidae (0.066) and a maximum value for Ac- ulations are protected by different conservation programs rocephalidae/Paridae (0.086). Myoglobin showed closer and the sampling process must be as noninvasive as pos- values: a maximum of 0.086 for the same families, but a sible. Previously, Ong and Vellayan (2008) showed the ap- minimum between Pycnonotidae/Phylloscopidae (0.051). plicability of CHD-based molecular sexing as a completely This last value is rather unexpected given that these two noninvasive method, using feathers as a source of DNA. families are taxonomically located in different clades. Di- The analysis of the sequence divergences for 7 taxa vergence between taxa based on cytb gene analysis shows representing 4 families of the superfamily Sylvioidea, plus high values that are uniformly distributed across taxa and an outgroup, Gallus gallus, indicates that the HKY dis- comparable to those of the CHD-Z and myo genes. tance for CHD-Z was between 0.000 and 0.290 (data not The evolution time of the CHD-Z gene indicates a shown). Divergences between cytochrome b (cytb) and higher substitution rate for transitions (2.536) and a lower myoglobin, intron 2 (myo), were higher than those of the rate for (0.511–1.000) (Table 2). The optimal CHD-Z gene. A graphical representation of the pairwise substitution model for CHD-Z gene evolution, selected by corrected sequence divergences (HKY distance) for CHD- the AIC algorithm in jModelTest (Guindon and Gascuel, Z, cytb, and myo is shown in Figure 2 and indicates a lin- 2003; Darriba et al., 2012), was TPM1uf + G (Kimura, ear relation among these three genes. The cytb sequence 1981), with 1.304 shape for 4 gamma categories. generally diverges much faster than CHD-Z (the ratio of The chi-square test of homogeneity of base frequencies cytb/CHD-Z is 1.41); the myoglobin gene also diverges across taxa was nearly significant (χ2 = 23.829, P = 0.068) faster than CHD-Z (the ratio of myo/CHD-Z is 1.15) for for CHD-Z marker across 39 variable sites within the in- the same set of taxa. Considering just the ingroup taxa group. Meanwhile, the chi-square test was clearly statisti- (the taxa from superfamily Sylvioidea), the myoglobin cally insignificant for the myo and cytb markers (P = 0.384 and CHD-Z genes have the same percentage of divergence for myo and P = 0.929 for cytb), suggesting a more stable (0.086%), much lower than cytb, which presents a value base composition among taxa. The base composition het- twice as large (0.178). erogeneity (RCV) confirms the chi-square test of homoge- Comparing the sequence divergence and the current neity, showing low levels of variability for the widely used classification within the superfamily Sylvioidea, the values markers cytb (RCV = 0.018) and myo (RCV = 0.025) and a are closely related for each taxonomic position. The spe- higher value for the new proposed marker CHD-Z (RCV = cies from the same genus revealed a divergence between 0.046). Phillips and Penny (2003) showed that 0.000 and 0.004 (genera Phylloscopus and Pycnonotus) for homogeneity within the ingroup and countable outgroup the CHD-Z gene compared to 0.003–0.102 for the cytb heterogeneity can equally affect the root position and the gene and 0.007–0.028 for the myo gene. CHD-Z interfam- tree topology. The high nucleotide homogeneity and the

Figure 2. HKY distance of cytochrome b vs. CHD-Z (left) and myoglobin vs. CHD-Z (right). The line represents the relationship of overall sequence divergences between two segments and indicates that the cytb and myo sequences are diverging faster than the CHD-Z gene sequence (cytb/CHD-Z = 1.41 and myo/CHD-Z = 1.15).

752 CIORPAC et al. / Turk J Zool

Table 2. Optimal substitution model selected based on AIC algorithm in jModelTest for the analyzed genes.

Gene AIC model selected R(AC) R(AG) R(AT) R(CG) R(CT) R(GT) ti/tv gamma p-inv CHD-Z TPM1uf+G 1.000 2.536 0.511 0.511 2.536 1.000 - 1.304 - cytb TVM+G 3.887 10.131 1.761 0.459 10.131 1.000 - 0.244 - myo HKY 1.914 - - - - - 1.914 - -

higher distance within individuals from four different eny of a single nuclear gene, in our case CHD-Z or myo, is families, observed for the cytb and myo markers, could insufficient and unreliable. explain the presence of uncertain relationships within the The mitochondrial sequence is a reliable tool for phylo- families of Sylvioidea. In contrast, the CHD-Z marker genetic relationships between supraspecific taxa, but usu- presents higher nucleotide heterogeneity, determined by ally it is not a good indicator of intra- or interspecific rela- the intron region, and a much lower distance within indi- tionships. Over time, combined data sets and nuclear and viduals, mainly determined by the exon region. mitochondrial markers have proven their usefulness in phy- Knowing the implications of molecular marker selec- logeny inference by increasing the resolution. For example, tion for phylogenetic analysis and the risk of obtaining dif- Pons et al. (2011) drew the green woodpecker’s phylogeog- ferent topologies using an improper marker, the topology raphy using mitochondrial marker cytb and another four given by the CHD-Z gene was tested along with those of nuclear markers (one of them being Z-linked), showing a the cytb and myo genes, the most common markers used clear distinction between the Iberian and European subspe- in bird phylogeny. The ML tree of a single nuclear gene re- cies of green woodpecker, soon confirmed by another study veals a wired topology for both analyses, CHD-Z and myo by Perktas et al. (2011) that used a different mitochondrial (Figure 3), probably caused by an improper tree algorithm marker over an increased number of individuals. Therefore, (ML usually gives good topology when the analyzed se- combined mitochondrial and nuclear data sets are man- quence has a high number of ) or by a short datory in order to eliminate any taxonomic uncertainties. distance between all studied taxa. Therefore, the phylog- Combining the cytochrome b sequence with the CHD-Z

Figure 3. The ML tree for myoglobin gene (in blue) and for CHD-Z gene (in red); both trees are scaled in substitutions and the length of each tree and different topologies indicate that both of them are improper for phylogeny using a single gene.

753 CIORPAC et al. / Turk J Zool gene and myoglobin, two data sets were obtained for a com- indicates the better model fit. The analyzed models’ accu- plete BI phylogenetic analysis to test the reliability of the racy is expressed as the difference between AICM values, CHD-Z gene in phylogeny. A topology comparison of the which means that the positive values (+2483.463) indicate MCC trees with cytochrome b highlights that all the MCC a better model relative fit of the cytb + CHD-Z tree com- trees have a similar topology, with differences given by the pared to the cytb + myo tree (Table 3). Furthermore, the total length of the branches (Figure 4), determined by the AICM values are confirmed by the tree mean posterior mutation rates specific for each gene. All these three MCC probability (cytb + CHD-Z 0.901, S.E. = 0.078; cytb 0.882, trees are scaled in substitutions and vary between 0.322 for S.E. = 0.0084; and cytb + myo 0.835, S.E. = 0.119), sug- cytb and 0.175 for myo and cytb. gesting a lower level of phylogenetic noise. Furthermore, The BI analysis showed a clock rate for CHD-Z-cytb the tree topology asymmetry was evaluated in order to as- genes (1.457), intermediary between myoglobin - cytb sess the combined data set constraining the species’ dif- (0.974) and the cytb gene (1.772) independently analyzed, ferent potential for speciation (Blum and Francois, 2005). meaning that the myo gene is changing faster through time The Colless index of tree imbalance (Colless, 1982) seems and can negatively affect the time-calibrated trees, giving to be unaffected by data sets, with a value of 0.466. Li et wrong divergence species time. A marginal likelihood es- al. (2007) described a good phylogenetic marker as a gene timation was performed for the tree models comparison with an intermediate substitution rate and high treeness by the AIC through Markov chain Monte Carlo (AICM) values, highlighting the importance of the optimal rate and algorithm, inferred using the method-of-moments estima- base composition stationarity in marker suitability. Even if tor (Baele et al., 2012) as a measure of accuracy of the mo- the CHD-Z base composition is less stationary than myo lecular clock models. The minimum value of the AICM, or cytb, the low number of variable sites with high base corresponding to the combined data set cytb + CHD-Z, composition heterogeneity seems to increase the fitness of

Figure 4. Maximum clade credibility trees for three data sets obtained by BI analysis in BEAST v1.8.0. MCC genealogy based on a Yule speciation process, summarized from the output of a Markov chain Monte Carlo chain run for 10 million iterations and sampled every 1000 iterations.

754 CIORPAC et al. / Turk J Zool

Table 3. Combined tree models compared by AICM (S.E. estimated from 1000 bootstrap replicates). The analysis was performed in Tracer v1.6 using log files from BEAST v1.85 analysis.

Combined tree model AICM S.E. cytb-z_chd-z.log.txt cytb-z_myo.log.txt cytb-z_chd-z.log.txt 7513.98 +/– 0.474 - 2483.463 cytb-z_myo.log.txt 9997.443 +/– 0.677 –2483.463 - this phylogenetic tool. Overall, our new proposed marker the relationship between closer and distant taxa. The CHD CHD-Z has the greatest phylogenetic signal-to-noise in- genes are reliable tools not just for sex determination in dex (treeness index: 0.371), and hence a lower potential birds, but also for population and phylogenetic analysis, for nonphylogenetic signals to influence phylogeny re- providing better results compared to the myoglobin gene, construction compared to myo (treeness 0.357), which a frequently used nuclear marker in bird analysis. is diverging faster than CHD-Z (the ratio of myo/CHD- Z is 1.15). Due to the coding and noncoding regions, the Acknowledgment CHD-Z marker is able to discriminate both close and dis- This work was supported by the project “Doctoral and tant sylvioid taxa, respectively. Post-doctoral programs of excellence for highly quali- In conclusion, the implications of the CHD gene for fied human resources training for research in the field of population analysis are substantial, offering a good image Life sciences, Environment and Earth Science”, Grant No. of population dynamics, sex distribution, and population POSDRU/159/1.5/S/133391, , cofinanced by the European trends. Additionally, the use of the CHD-Z gene in phylo- Social Fund within the Sectorial Operational Program genic reconstruction gives the most effective results, hav- Human Resources Development 2007–2013, University of ing a moderate clock rate and being capable of revealing Bucharest, Romania.

References

Akaike H (1974). A new look at the statistical model identification. Boardman PE, Sanz-Ezquerro J, Overton IM, Burt DW, Bosch IEEE T Automat Contr 19: 716-723. E, Fong WT, Tickle C, Brown WR, Wilson SA, Hubbard SJ (2002). A comprehensive collection of chicken cDNAs. Curr Alstrom P, Ericson PG, Olsson U, Sundberg P (2006). Phylogeny and Biol 12: 1965-1969. classification of the avian superfamily Sylvioidea. Mol Phylo- genet Evol 38: 381-397. Bouckaert RR (2010). DensiTree: making sense of sets of phyloge- netic trees. Bioinformatics 26: 1372-1373. An J, Lee MY, Min MS, Lee MH, Lee H (2007). A molecular genetic approach for species identification of and sex deter- Bradley RD, Baker RJ (2001). A test of the genetic species concept: mination of birds in a forensic case of poaching from South cytochrome-b sequences and mammals. J 82: 960- Korea. Forensic Sci Int 167: 59-61. 973. Baele G, Lemey P, Bedford T, Rambaut A, Suchard MA, Alekseyenko Burbrink FT, Lawson R, Slowinski JB (2000). Mitochondrial DNA AV (2012). Improving the accuracy of demographic and mo- phylogeography of the polytypic North American rat snake lecular clock model comparison while accommodating phylo- (Elaphe obsoleta): a critique of the subspecies concept. Evolu- genetic uncertainty. Mol Biol Evol 29: 2157-2167. tion 54: 2107-2118. Baker JM, Lopez-Medrano E, Navarro-Siguenza AG, Rojas-Soto OR, Cagnon C, Lauga B, Hemery G, Mouches C (2004). Phylogeographic Omland KE (2003). Recent speciation in the orchard oriole differentiation of storm petrels (Hydrobates pelagicus) based group: divergence of Icterus spurius spurius and Icterus spurius on cytochrome b mitochondrial DNA variation. Mar Biol 145: fuertesi. Auk 120: 848-859. 1257-1264. Bensch S, Irwin DE, Irwin JH, Kvist L, Akesson S (2006). Conflicting Chang HW, Cheng CA, Gu DL, Chang CC, Su SH, Wen CH, Chou patterns of mitochondrial and nuclear DNA diversity in Phyl- YC, Chou TC, Yao CT, Tsai CL et al. (2008). High-throughput loscopus warblers. Mol Ecol 15: 161-171. avian molecular sexing by SYBR green-based real-time PCR combined with melting curve analysis. BMC Biotechnol 8: 12. Blum MG, Francois O (2005). On statistical tests of phylogenetic tree imbalance: the Sackin and other indices revisited. Math Biosci Chang HW, Chou YC, Su YF, Cheng CA, Yao CT, Tsai CL, Lee HC, 195: 141-153. Wen CH, Cheng CC (2010). Molecular phylogeny of the Pyc- nonotus sinensis and Pycnonotus taivanus in Taiwan based on sequence variations of nuclear CHD and mitochondrial cyto- chrome b genes. Biochem Syst Ecol 38: 195-201.

755 CIORPAC et al. / Turk J Zool

Colless D (1982). Review of phylogenetics: the theory and practice of Heidrich P, Amengual J, Wink M (1998). Phylogenetic relationships phylogenetic systematics. Syst Zool 31: 100-104. in Mediterranean and North Atlantic shearwaters (Aves: Pro- cellariidae) based on nucleotide sequences of mtDNA. Bio- Darriba D, Taboada GL, Doallo R, Posada D (2012). jModelTest chem Syst Ecol 26: 145-170. 2: more models, new heuristics and parallel computing. Nat Methods 9: 772. Helbig AJ, Seibold I, Martens J, Wink M (1995). Genetic differentia- tion and phylogenetic relationships of Bonelli’s Warbler Phyl- Drummond AJ, Ho SY, Phillips MJ, Rambaut A (2006). Relaxed phy- loscopus bonetti and Green Warbler P. niti du s . J Avian Biol 26: logenetics and dating with confidence. PLoS Biol 4: e88. 139. Drummond AJ, Suchard MA, Xie D, Rambaut A (2012). Bayesian Hung GC, Chilton NB, Beveridge I, Zhu XQ, Lichtenfels JR, Gasser phylogenetics with BEAUti and the BEAST 1.7. Mol Biol Evol RB (1999). Molecular evidence for cryptic species within Cyli- 29: 1969-1973. costephanus minutus (Nematoda: Strongylidae). Int J Parasitol Dubiec A, Zagalska-Neubauer M (2006). Molecular techniques for 29: 285-291. sex identification in birds. Biological Letters 43: 3-12. Johansson US, Ekman J, Bowie RC, Halvarsson P, Ohlson JI, Price Ellegren H (1996). First gene on the avian W chromosome (CHD) TD, Ericson PG (2013). A complete multilocus species phylog- provides a tag for universal sexing of non-ratite birds. Proc Biol eny of the tits and chickadees (Aves: Paridae). Mol Phylogenet Sci 263: 1635-1641. Evol 69: 852-860. Fregin S, Haase M, Olsson U, Alstrom P (2009). Multi- phylog- Johnson NK, Cicero C (2004). New mitochondrial DNA data affirm eny of the family Acrocephalidae (Aves: Passeriformes)--the the importance of Pleistocene speciation in North American traditional taxonomy overthrown. Mol Phylogenet Evol 52: birds. Evolution 58: 1122-1130. 866-878. Kahn NW, St John J, Quinn TW (1998). Chromosome-specific in- Fridolfsson AK, Ellegren H (1999). A simple and universal method tron size differences in the avian CHD gene provide an efficient for molecular sexing of non-ratite birds. J Avian Biol 30: 116- method for sex identification in birds. Auk 115: 1074-1078. 121. Kimura M (1981). Estimation of Evolutionary Distances between Fuchs J, Fjeldså J, Bowie RC, Voelker G, Pasquet E (2006). The Afri- Homologous Nucleotide-Sequences. P Natl Acad Sci-Biol 78: can warbler genus Hyliota as a lost lineage in the Oscine song- 454-458. bird tree: molecular support for an African origin of the Pas- Lei X, Yin Z, Lian Z, Chen C, Dai C, Kristin A, Lei F (2010). Phyloge- serida. Mol Phylogenet Evol 39: 186-197. netic relationships of some Sylviidae species based on complete Gernhard T (2008). The conditioned reconstructed process. J Theor mtDNA cyt b and partial COI sequence data. Chinese Birds 1: Biol 253: 769-778. 175-187. Gill FB, Slikas B, Sheldon FH, Fleischer R (2005). Phylogeny of tit- Li C, Ortí G, Zhang G, Lu G (2007). A practical approach to phyloge- mice (Paridae): II. Species relationships based on sequences of nomics: the phylogeny of ray-finned fish (Actinopterygii) as a the mitochondrial cytochrome-b gene. Auk 122: 121-143. case study. BMC Evol Biol 7: 1. Gouy M, Guindon S, Gascuel O (2010). SeaView version 4: A multi- Lin CP, Danforth BN (2004). How do insect nuclear and mitochon- platform graphical user interface for sequence alignment and drial gene substitution patterns differ? Insights from Bayesian phylogenetic tree building. Mol Biol Evol 27: 221-224. analyses of combined datasets. Mol Phylogenet Evol 30: 686- Griffiths R, Double MC, Orr K, Dawson RJ (1998). A DNA test to sex 702. most birds. Mol Ecol 7: 1071-1075. Lohman DJ, Ingram KK, Prawiradilaga DM, Winker K, Sheldon FH, Griffiths R, Tlwarl B (1995). Sex of the last wild Spix’s macaw. Nature Moyle RG, Ng PKL, Ong PS, Wang LK, Braile TM et al. (2010). 375: 454. Cryptic genetic diversity in “widespread” Southeast Asian bird species suggests that Philippine avian endemism is gravely un- Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Gascuel derestimated. Biol Conserv 143: 1885-1890. O (2010). New algorithms and methods to estimate maximum- likelihood phylogenies: assessing the performance of PhyML Miller M, Pfeiffer W, Schwartz T (2010). Creating the CIPRES Sci- 3.0. Syst Biol 59: 307-321. ence Gateway for inference of large phylogenetic trees. In: Gateway Computing Environments Workshop (GCE), 2010. Guindon S, Gascuel O (2003). A simple, fast, and accurate algorithm New York, NY, USA: IEEE, pp 1-8. to estimate large phylogenies by maximum likelihood. Syst Biol 52: 696-704. Newman CE, Feinberg JA, Rissler LJ, Burger J, Shaffer HB (2012). A new species of leopard frog (Anura: Ranidae) from the urban Harvey MG, Bonter DN, Stenzler LM, Lovette IJ (2006). A compari- northeastern US. Mol Phylogenet Evol 63: 445-455. son of plucked feathers versus blood samples as DNA sources for molecular sexing. J Field Ornithol 77: 136-140. Nishibori M, Shimogiri T, Hayashi T, Yasue H (2005). Molecular evi- dence for hybridization of species in the genus Gallus except Hasegawa M, Kishino H, Yano T (1985). Dating of the human-ape for Gallus varius. Anim Genet 36: 367-375. splitting by a molecular clock of mitochondrial DNA. J Mol Evol 22: 160-174.

756 CIORPAC et al. / Turk J Zool

Olsson U, Alström P, Ericson PG, Sundberg P (2005). Non-mono- Simic R, Lindstrom DL, Tran HG, Roinick KL, Costa PJ, Johnson phyletic taxa and cryptic species - Evidence from a molecular AD, Hartzog GA, Arndt KM (2003). Chromatin remodeling phylogeny of leaf-warblers (Phylloscopus, Aves). Mol Phylo- protein CHD1 interacts with transcription elongation factors genet Evol 36: 261-276. and localizes to transcribed genes. EMBO J 22: 1846-1856. Ong AH, Vellayan S (2008). An evaluation of CHD-specific primer Tamura K, Stecher G, Peterson D, Filipski A, Kumar S (2013). sets for sex typing of birds from feathers. Zoo Biology 27: 62- MEGA6: Molecular Evolutionary Genetics Analysis version 69. 6.0. Mol Biol Evol 30: 2725-2729. Parkin DT, Collinson JM, Helbig AJ, Knox AG, Sangster G, Svensson Vucicevic M, Stevanov-Pavlovic M, Stevanovic J, Bosnjak J, Gajic B, L (2004). Species limits in Acrocephalus and Hippolais warblers Aleksic N, Stanimirovic Z (2013). Sex determination in 58 bird from the Western Palearctic. Brit Birds 97: 276-299. species and evaluation of CHD gene as a universal molecular marker in bird sexing. Zoo Biology 32: 269-276. Perktas U, Barrowclough GF, Groth JG (2011). Phylogeography and species limits in the green woodpecker complex (Aves: Pici- Wesson DM, McLain DK, Oliver JH, Piesman J, Collins FH (1993). dae): multiple Pleistocene refugia and range expansion across Investigation of the validity of species status of Ixodes-Dam- Europe and the Near East. Biol J Linn Soc 104: 710-723. mini (Acari, Ixodidae) using rDNA. P Natl Acad Sci USA 90: 10221-10225. Phillips MJ, Penny D (2003). The root of the mammalian tree in- ferred from whole mitochondrial . Mol Phylogenet Yule GU (1925). A mathematical theory of evolution, based on the Evol 28: 171-185. conclusions of Dr. JC Willis, FRS. Philos T R Soc Lon B 213: 21-87. Pons JM, Olioso G, Cruaud C, Fuchs J (2011). Phylogeography of the Eurasian green woodpecker (Picus viridis). J Biogeogr 38: Zhang S, Yang L, Yang XJ, Yang JX (2007). Molecular phylogeny of 311-325. the yuhinas (Sylviidae: Yuhina): a paraphyletic group of bab- blers including Zosterops and Philippine Stachyris. J Ornithol Price T, Birch GL (1996). Repeated evolution of sexual color dimor- 148: 417-426. phism in passerine birds. The Auk 842-848. Zuccon D, Ericson PGP (2010). The phylogenetic position of the Seutin G, White BN, Boag PT (1991). Preservation of avian blood Black-collared Bulbul Neolestes torquatus. Ibis 152: 386-392. and tissue samples for DNA analyses. Can J Zool 69: 82-90.

757